Live Exposure Defense: From CVE to Confirmed Exposure in 12 Hours – See more

Back
New CVE Detected

CVE-2026-77521 – Unauthenticated Remote Command Execution (via Prompt Injection) – MaxKB prior to 2.

Be the first to know when new zero-days emerge:

Subscribe to Email Alerts
Subscribe to RSS Feed

Summary

CVE-2026-77521 is a critical OS command injection vulnerability in MaxKB, an open-source AI assistant platform, affecting versions prior to 2.10.5-lts. MaxKB assistants configured with a tool, MCP tool, skill, or sub-application route chat through an agent backend that exposes an unrestricted shell "execute" tool without requiring human approval, allowing untrusted chat input or ingested content to trigger arbitrary command execution. The flaw carries the maximum CVSS score of 10.0 due to its network exploitability, lack of required privileges, and full impact on confidentiality, integrity, and availability.

Technical details

  • Root cause: MaxKB assistants with an attached tool, MCP tool, skill, or sub-application use a SandboxShellBackend that automatically exposes an execute shell tool alongside other file/shell utilities; this tool is never excluded from the assistant’s available tools and is omitted from the set of actions requiring human ("interrupt_on") approval before execution.
  • Trigger conditions: An attacker who can inject content into a chat session or into data later ingested by the assistant (indirect prompt injection) can instruct the underlying model to invoke the execute tool, causing shell command execution without any operator confirmation.
  • Deployment-specific impact: In source/bare-metal deployments where the MAXKB_SANDBOX setting is not enabled (it is off by default), commands run directly on the host as the application user via a shell-invoked subprocess call. In the official container image, neither Dockerfile defines a non-root USER, and the sandbox wrapper itself invokes a shell using a string command, so shell metacharacters and redirection can be used to escape the intended sandbox restrictions and execute as root.
  • Attack vector: Network-based, no authentication or user interaction beyond the ability to submit chat content or ingested data required (AV:N/AC:L/PR:N/UI:N).
  • Impact: Full compromise of confidentiality, integrity, and availability of the host running MaxKB, including in multi-tenant or publicly exposed chatbot deployments.

Affected software

  • MaxKB (1Panel-dev) versions v2.10.3-lts and earlier, and all prior releases up to and including that version
  • Applies to assistants that have a tool, MCP tool, skill, or sub-application attached, which routes chat through the vulnerable shell backend

Severity

  • CVSS v3.1 Base Score: 10.0 (Critical)
  • Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H

Mitigation and recommended actions

  • Immediate: Upgrade MaxKB to v2.10.5-lts or later, which fixes the unrestricted shell tool exposure.
  • If immediate patching is not possible:
    • Disable or remove tools, MCP tools, skills, and sub-applications from assistants that do not strictly require shell/file execution capability.
    • Ensure MAXKB_SANDBOX is enabled for all deployments so shell execution is confined to a sandboxed context.
    • For containerized deployments, avoid running the MaxKB container as root; apply a non-root USER directive where possible.
    • Restrict network exposure of MaxKB chat interfaces to trusted users only, and avoid exposing assistants with attached tools to public or untrusted input until patched.

References

Are you exposed?

Get a free report of your organization’s exposure to this CVE and threat

  • Mapping of all Assets with this Technology
  • Identification of potentially exposed assets to this CVE
  • Confirmation of verified exploitable assets
Get Exposure Report

How IONIX’s External Exposure Management Platform Detects and Validates
Zero-Days to Shrink MTTR

1

Map your entire attack surface (continously)

IONIX uses multi-factor discovery methods, including DNS analysis, certificate mapping, metadata inspection, and more, to automatically map every internet-facing asset across your environment. This includes cloud instances, third-party platforms, shadow IT, and even forgotten infrastructure that traditional tools miss.

2

Monitor for new CVEs

Dozens of threat intel feeds using agentic technology are continuously analyzed to detect the appearance of proof-of-concept code, exploit kits, and indicators of active targeting. IONIX goes further by applying AI to proactively evaluate whether emerging vulnerabilities are likely to be exploited, even before PoCs go public.

3

Identify Potential External Exposures

Not all CVEs matter. IONIX filters vulnerabilities by asking attacker-centric questions: Can it be reached from the internet? Does it require authentication? Is it being exploited in the wild? This dramatically reduces noise and focuses teams on threats that can actually be weaponized.

4

Create Safe, Scalable Exploit Validations

IONIX transforms real-world PoCs into safe, non-intrusive test payloads that can be run in production environments without disruption. These simulations are precisely targeted to the systems that are vulnerable, ensuring rapid validation without unnecessary load.

5

Execute Exploit Validations

By combining context about software stack, versioning, exposure status, and reachability, IONIX ensures that only the right payloads are executed against the right assets, maximizing efficiency and minimizing risk.

6

Drive Fast and Actionable Remediation

Results are routed through integrations with ticketing, SOAR, and SIEM tools. Issues are written in plain language, bundled into remediation clusters, and prioritized based on asset criticality, exploitability, and blast radius. This shortens mean time to remediation (MTTR) and empowers teams to act with confidence.

Are you exposed?

Get a free report of your organization’s exposure to this CVE and threat

  • Mapping of all Assets with this Technology
  • Identification of potentially exposed assets to this CVE
  • Confirmation of verified exploitable assets
Get Exposure Report
Close

Subscribe to Threat Center RSS

Copy/paste the link below into your preferred RSS reader or follow these instructions to subscribe to Slack alerts.

Open RSS feedSlack setup instructions

Close

Get Real-Time CVE Alerts to Your Email

Be the first to know when new zero-days emerge